How simple probability predicts humanity's remaining lifespan
Demographers estimate that roughly 117 billion humans have lived throughout history. Assuming you are a typical human placed at a random position in human history, standard statistical sampling suggests that total human history will likely cap out around 200 billion people. If population growth continues, the Doomsday Argument warns that humanity has a surprisingly high probability of facing extinction or catastrophe within the next few centuries.
The Copernican Shift in Human History
In the sixteenth century, Nicolaus Copernicus displaced humanity from the physical center of the universe by showing that Earth revolves around the Sun. Centuries later, astrophysicists and philosophers realized that this same principle of mediocrity—the idea that an observer should not assume they occupy a uniquely privileged position—could be applied not just to space, but to time. If you do not occupy a special point in the history of the universe, you are also unlikely to be occupying a uniquely special epoch in the lifespan of your own species.
This insight forms the foundation of the Doomsday Argument, an anthropic thought experiment independently developed by astrophysicist Brandon Carter in 1983 and later popularized by philosopher John Leslie. The argument proposes that purely by considering our position in the chronological order of human births, we can make probabilistic estimates about how many total humans will ever exist. Rather than relying on geopolitical forecasts or biological modeling, it uses basic probability to question whether humanity has a vast cosmic future ahead.
How the Math Limits Our Future
The mathematical engine behind the Doomsday Argument relies on what philosopher Nick Bostrom called the Self-Sampling Assumption: you should reason as if you are a random sample drawn uniformly from the set of all observers who will ever live. Demographers estimate that roughly 100 to 115 billion humans have been born throughout history up to the present moment. If your birth rank is denoted as n, and the total number of humans who will ever be born across all of history is N, your position n is simply a random number between 1 and N.
In any random draw between 1 and N, the probability that your draw lands in the first 5 percent of the total sequence is exactly 5 percent. Stated conversely, there is a 95 percent probability that you were born somewhere after the first 5 percent of all humans. If your birth rank n is around 115 billion, setting that value at the 5 percent threshold implies that total human births N are unlikely to exceed 2.3 trillion. If global birth rates remain near modern levels, that remaining population budget would be exhausted within millennia or even centuries, rather than the millions of years typically envisioned for an interstellar civilization.
The Delta t Approach and Physical Durations
In 1993, astrophysicist J. Richard Gott III published an alternative formulation in the journal Nature, often referred to as the Delta t argument. Gott illustrated the concept through his visit to the Berlin Wall in 1969. Assuming his visit occurred at a random point in the wall's total existence, he calculated with 75 percent confidence that the wall would stand for at least one-third of the time it had already stood, but no more than three times its current age. The wall fell twenty years later, in 1989, fitting well inside his predicted interval.
Gott extended this frequentist reasoning from physical structures and Broadway show runs to the survival of intelligent species. If Homo sapiens has existed for roughly 200,000 years, applying a 95 percent confidence interval suggests humanity's total future duration is likely between approximately 5,100 years and 7.8 million years. While 7.8 million years sounds extensive, it falls far short of astronomical timescales and drastically lowers the probability that human descendants will populate the galaxy for billions of years.
The Reference Class Problem
Despite its mathematical simplicity, the Doomsday Argument faces intense philosophical scrutiny. One of the most stubborn vulnerabilities is known as the reference class problem. To compute your relative position among all observers, you must define who qualifies as a member of your reference class. Does the category include only anatomically modern Homo sapiens, or does it include Neanderthals, early hominids, genetically modified post-humans, uploaded digital minds, and intelligent extraterrestrials?
If future human civilization transitions into vast populations of artificial intelligences or vastly modified biological beings, and those beings belong to the same reference class as modern humans, the total number of observers N becomes extraordinarily large. Under that broader definition, being born among the first 100 billion biological humans would indeed place an observer in a tiny fraction of the total population, undermining the assumption that our birth rank represents a typical sample.
The Self-Indication Assumption
Another major challenge comes from the Self-Indication Assumption, formulated by physicists and philosophers including Dennis Dieks and Ken Olum. This principle suggests that the very fact of your existence provides evidence that the universe contains a large number of observers rather than a small number. In a hypothetical universe where trillions of humans are born, the probability of any individual being brought into existence is vastly higher than in a universe where humanity goes extinct after only 200 billion births.
When this conditional probability of existing is factored into a Bayesian update, it mathematically balances out the Doomsday Argument's downward pressure on humanity's lifespan. Under the Self-Indication Assumption, observing that you exist at a particular birth rank does not systematically favor an early extinction, because universes with massive total populations generate far more opportunities for observers to experience existence in the first place.
A Mirror for Existential Risk
While the Doomsday Argument remains hotly contested among statisticians and philosophers, it played a central historical role in shaping modern discussions of existential risk and observation selection effects. By formalizing how selection biases influence scientific reasoning, it forced researchers to confront the challenges of self-locating belief—how an observer's position inside a system changes the inferences they can make about the system as a whole.
Whether viewed as a genuine warning of impending catastrophe or as a subtle paradox exposing the limits of anthropic probability, the argument serves as a reminder of humanity's precarious situation. It demonstrates that our assumptions about how ordinary or exceptional we are have profound, mathematically rigorous consequences for how we envision the distant future.
Key takeaways
•The Doomsday Argument uses the Copernican principle to suggest that, assuming you are a typical human, your birth rank implies a limited total number of future humans.
•J. Richard Gott's Delta t formulation applies identical probabilistic reasoning to duration, predicting that humanity's future lifespan is bounded by its past duration.
•The argument depends heavily on the 'reference class'—the exact definition of which beings count as fellow observers across history and the future.
•Counterarguments like the Self-Indication Assumption propose that existing at all biases probability toward universes with larger total populations, countering the doomsday prediction.